US 7183100 B2
A bacterial protein which converts 2-methyl citrate to 2-methyl isocitrate is a previously unknown target for antibacterial agents. The protein of this activity is associated with mucoid bacteria and inhibitors of production or activity of this protein in combination with propionic acid mitigate the virulence of these bacteria.
1. An isolated and purified recombinant protein which comprises the amino acid sequence of SEQ ID NO: 6.
2. An isolated and purified recombinant protein which consists of an amino acid sequence which is at least 98% sequence homologous to the amino acid sequence of aconitase SEQ ID NO: 6 over its entire length and which exhibits enzymatic activity to convert 2-methyl citrate to 2-methyl isocitrate.
3. A method to identify a test compound that, in combination with propionic acid, inhibits the virulence of mucoid bacteria, which method comprises assessing the activity of the protein of
4. The method of
5. A method to identify a test compound that, in combination with propionic acid, inhibits the virulence of mucoid bacteria, which method comprises assessing the activity of the protein of
The present application is related to and claims priority to U.S. Patent Application Ser. No. 60/459,885, filed Apr. 1, 2003 now abandoned. The contents of this application is incorporated herein by reference.
The invention relates to methods to control the growth and virulence of mucoid bacteria and to regulate their production of exopolysaccharide biofilms. The invention also is directed to methods to screen for useful antibiotics. Such screening methods employ a novel aconitase whose properties have heretofore been unknown.
Several species of bacteria are able to secrete exopolysaccharides or alginates that are essential for virulence as the exudate provides a mechanism for adherence and colonization. One particularly important example of such bacteria is Pseudomonas aeruginosa, a common soil bacterium which inhabits individuals generally, but is particularly destructive in subjects with cystic fibrosis. Cystic fibrosis is an autosomal recessive genetic disorder linked to dysfunctional CFTR chloride channels on cell surfaces. It is characterized by production of thick mucus which prevents clearance of bacteria, resulting in chronic infection and inflammation. Because P. aeruginosa produces biofilm in the lungs and digestive systems of these subjects, and the subjects are unable to clear this biofilm, P. aeruginosa infection is a major cause of death among such individuals.
Many bacteria exhibit mucoid (resembling mucus) phenotype as a response to their growth environment. The mucoidy is generated by bacterial production of extracellular polysaccharides (exopolysaccharide or EPS). Various EPS molecules that include frucose, rugose and glucose residues have been characterized. Examples of bacteria producing mucoid phenotype include alginate producing Pseudomonas and Azotobacter species (i.e., P. aeruginosa, Azotobacter vinelandii), rugose producing Vibrio species (i.e., Vibrio cholerae), xanthan producing Xanthomonas species (i.e., Xanthomonas campestris), gellan producing Sphingomonas species (i.e., S. paucimobilis), curdlan-type EPS producing Cellulomonas, Alcaligenes and Agrobacterium species (i.e., Cellulomonas flavigena, Alcalifenes faecalis) and Shewanella, Bordetella and Streptococcus species producing various uncharacterized EPS, among others. In fact, under unfavorable growth conditions, many bacteria can switch to a mucoid phenotype to resist the environmental stress and adapt to unfavored conditions. Biofilms represent a typical structured adaptation environment in which many bacteria co-exist and secrete extracellular polysaccharides which aid them to stick to surfaces for growth and colonization, provide a protective barrier around them and adapt to their environment in a microbial community. The exopolysaccharides produced by mucoid Pseudomonas that occupy the lungs of cystic fibrosis patients are generally alginates which are O-glycosyl linked D-mannuronate and L-guluronate residues.
It is understood that, in order to produce these biofilms, the essential components required must be made available by the metabolic system of the bacterium. The present invention provides means to disrupt this ability by disabling an essential step in this metabolic sequence.
The present inventors have discovered that mucoid bacteria contain an aconitase, designated herein that encoded by acnC which catalyzes the conversion of 2-methyl citrate to 2-methyl isocitrate. This aconitase, which has an activity different from aconitases previously known, is required for the clearance of propionate; propionic acid is a known toxic agent for mucoid bacteria as described in PCT publication WO 01/30997, the disclosure of which is incorporated herein by reference. Accordingly, disruption of the activity or production of acnC protein along with the administration of propionic acid, or of compounds which generate propionic acid, has a deleterious effect on mucoid producing bacteria. This effect resides, in large part, in inhibiting the production of the biofilm and thus inhibiting the ability of the bacterium to survive in its environment.
Thus, in one aspect, the invention is directed to a method to mitigate the virulence of a mucoid bacterial culture or infection, which method comprises contacting the bacteria contained in said culture or infection with an effective amount of propionic acid or a substance which generates propionic acid in combination with effecting inhibition of the production or activity of acnC protein.
In another aspect, the invention is directed to a method to screen for compounds that enhance the toxicity of propionic acid to exopolysaccharide-producing bacteria, which method comprises assessing the ability of candidate compounds to inhibit the activity of acnC protein. This method comprises determining the conversion of 2-methyl citrate to 2-methyl isocitrate in the presence of acnC protein and testing this in the presence and absence of a candidate compound. Compounds whose presence reduces the level of 2-methyl isocitrate produced, or which are otherwise shown to inhibit acnC protein, are identified as useful in enhancing the toxicity of propionic acid or its precursors.
In a third aspect, the invention is directed to a composition of matter which comprises an isolated form of acnC protein, compositions which comprise recombinant materials for its production and methods for producing acnC protein using the recombinant materials. The invention also is directed to antisense or triplex forming nucleic acid molecules and other inhibitors for the production of acnC protein.
While many currently used antibiotics and drugs target the ability of bacteria to grow, they do not necessarily reduce the ability of these bacteria to infect the host, to adapt, and to produce virulence factors. The present invention, by elucidating the function and structure of a protein product and its encoding gene that is involved in virulence and metabolic adaptation, provides an entirely new target for the design and development of new anti-infectives, antibacterial compounds, and biofilm control agents. As this is a new target for antibacterial drugs, resistance to such drugs has not developed.
This target is exemplified herein by the isolation and manipulation of a gene from Pseudomonas aeruginosa designated aconitase C (acnC). Disruption of acnC completely abolishes bacterial growth in the presence of propionic acid and results in a significant reduction in the virulence factors associated with P. aeruginosa, including the production of biofilms. Hence, inhibition of the production of this protein or inhibition of its activity will attenuate microbial virulence in the presence of propionic acid or a material which generates it. No eukaryotic counterpart to aconitase C is known.
It is demonstrated herein that disruption of aconitase C activity destroys the ability of mucoid bacteria to grow in the presence of propionic acid. Accordingly, it is clear that inhibitors of this activity, either those which inhibit the production of the protein or those which inhibit the activity of the protein itself will be useful in making mucoid bacteria more susceptible to propionic acid and to agents which generate propionic acid.
Previously described aconitase activities catalyze the dehydration of citric acid to cis aconitate. The presently isolated aconitase C is capable of this activity as well, but has the additional feature of catalyzing the conversion of 2-methyl citrate to 2-methyl isocitrate as shown in
“Aconitase C” and “acnC” are used interchangeably and refer to any nucleotide sequence encoding the protein with enzymatic activity, the protein itself, and the gene locus which results in the production of the protein. In order to be defined as “aconitase C” or “acnC” the protein must exhibit the ability to convert 2-methyl citrate to 2-methyl isocitrate. This activity can readily be verified using routine enzymatic assays. In some drawings and text herein, the corresponding materials are also labeled “acaB.” In Pseudomonas, the acnC encoding gene is found downstream in the propionate operon from prpC and upstream of prpD.
One assay is analogous to that employed for determining levels of aconitase activity known in the prior art, based on the sequence of reactions shown below:
As the aconitase C of the present invention converts 2-methyl citrate to 2-methyl isocitrate, and as 2-methyl isocitrate is also oxidized with NADP+ in the presence of isocitrate dehydrogenase, a similar assay that couples spectrophotometric measurement of NADPH production at 340 nm can be used to determine levels of aconitase C of the present invention.
The foregoing, and any other appropriate assay for the conversion of 2-methyl citrate to 2-methyl isocitrate, can be used as a screening assay to identify compounds that will be useful in modifying the virulence of mucoid bacteria. Compounds which inhibit this activity will be useful in this regard. Thus, the activity of a preparation of purified and isolated aconitase C protein, recombinantly produced aconitase C protein, or even an impure preparation of aconitase C protein is tested for this conversion activity in both the presence and absence of candidate compound. Compounds whose presence results in a decrease in this activity are identified as useful compounds for reducing the virulence of mucoid bacteria. As a preliminary screen, the traditional aconitase activity assay shown above can be used as a surrogate in identifying compounds with the desired activity. However, assays using the aconitase C protein per se are preferred.
The screening assay using aconitase C protein per se is facilitated by virtue of the availability of recombinant materials for production of the required aconitase C protein. Described herein, as illustrative, is the amino acid sequence of the aconitase C gene derived from P. aeruginosa. Homologous proteins from other prokaryotic organisms could also be used, and can readily be retrieved using standard techniques with the information contained in the P. aeruginosa gene as a guide. Thus, the protein encoded by the gene in the P. aeruginosa strain 01 comprises 869 amino acids and the nucleotide sequence and deduced amino acid sequence are shown in
Proteins with homology to the PA01 aconitase C protein also share the ability to convert 2-methyl citrate to 2-methyl isocitrate. Thus, included within the scope of the invention are proteins which exhibit at least 85%, preferably 90%, preferably 95%, and more preferably 98% homology over the entire sequence to the sequence shown in
Similarly, proteins encoded by a nucleotide sequence which hybridizes under stringent conditions to a nucleotide sequence encoding the amino acid sequence of the PA01 aconitase C protein are also included within the scope of the invention. The stringency of hybridization is defined by the wash conditions subsequent to the hybridization itself and “stringent” conditions are defined as washing in 0.1% SSC at 65° C.
Thus, proteins which are within the scope of the present invention may be defined in terms of their ability to convert 2-methyl citrate to 2-methyl isocitrate in combination (1) with specified homology to the acnC sequence set forth in
While nucleic acids which encode aconitase C proteins can be defined in terms of nucleotide sequences degenerate with that set forth in
As stated above, one approach to modulating the virulence of mucoid bacteria comprises contacting such bacteria with a compound which inhibits the aconitase C activity in combination with a source of propionate. Such compounds can be identified through the screening assay described above, or may already be known to block aconitase C activity by virtue of their ability to bind the aconitase C protein. Thus, antibodies or other specific binding partners for the aconitase C proteins of the invention may be employed. “Antibodies” include, in addition to immunoglobulins in general, immunoreactive portions, such as the F(ab) or F(ab′) or F(ab′)2 fragments; antibodies may also be prepared as single-chain forms—i.e., scFv antibodies. Various ways to manipulate antibodies for particular purposes are also well known; thus, included within the invention are humanized forms of antibodies of the invention or antibodies which are modified to correspond to the species to which they may be administered. Other specific binding partners include aptamers—i.e., nucleic acids which optionally have been selected through known rounds of selection, for example, for specific binding to proteins with aconitase C activity. Such aptamers may be “traditional” nucleic acids or modified forms thereof, such as peptide nucleic acids. Thus, aconitase C activity may be inhibited in a variety of ways, including direct binding of the protein by antibodies or aptamers, and by compounds which have been shown to inhibit the activity empirically.
In addition to use of compounds which inhibit aconitase C activity, methods to inhibit the production of the aconitase C protein itself may also be employed. Such known methods include use of antisense constructs and formation of a triple helix at a critical position in the gene. In these methods, of course, the native, non-degenerate nucleotide sequence must be targeted. Thus, suitable targets for triplex formation or antisense inhibition include nucleotide sequences which encode aconitase C activity and which have at least 85%, preferably 90% homology, preferably 95% homology and more preferably 98% homology to the nucleotide sequence shown to encode acnC in
In the method of the invention, the mucoid bacteria for which virulence is sought to be modulated is treated both with materials which inhibit production or activity of aconitase C and with a source of propionate. The propionate source may be propionic acid itself or a fatty acid with an odd number of carbon atoms in the chain which generates propionic acid metabolically. Other materials known to be metabolized to propionic acid, such as ibuprofen, could also be substituted. Thus, the propionic acid portion of the treatment is contributed by any compound which is itself propionic acid or generates propionic acid in situ.
The propionic acid source and the means for modulating aconitase C activity (which means include direct inhibition of activity and inhibition of production of this protein) are supplied either simultaneously or sequentially to the mucoid bacterium target. If the targeted bacterium is present in an in vitro environment, e.g., in a foodstuff or other composition to be decontaminated, the propionic acid source and the modulator of acnC activity can be supplied directly to this material. Alternatively, the offending bacterium may have infected an organism, in which case the appropriate materials are supplied to the organism per se. Thus, for example, if the targeted bacterium has infected an animal, such as a mammal or avian host, the mammal or avian host is provided a propionic acid source and a modulator of acnC. Such administration can be by any traditional means and the materials formulated appropriately to their nature. Thus, administration may be by injection, transmucosal, transdermal, topical, local, systemic, oral, or in a variety of paradigms well known to practitioners.
When the combination of a propionic acid generating compound and inhibitor or modular of acnC is used to treat a subject, the choice of propionic acid generating compound is made appropriate to lack of toxicity and ability of the treated subject to metabolize the precursor to the desired product; thus, for example, propionic acid per se would not be used in mammalian subjects. Suitable formulations are also provided as is known in the art for effective routes of administration.
By “treat” bacterial infection is meant any positive change with regard to the health of the subject related to the underlying infection, not necessarily a complete recovery. Thus, reduction of the mucoid production by the bacteria, amelioration of symptoms, slowing the progression of bacterial growth, and the like, are all within the scope of “treating.”
The following examples are intended to illustrate but not to limit the invention.
The genome of P. aeruginosa wildtype strain 01, a mucoid producing strain, designated herein PA01 was subjected to PCR to obtain a 900 base pair aconitase C encoding fragment. The forward primer was GTNGGNACNGAYTCNCAYACN (SEQ ID NO:1) and the backward primer was NCKNCCYTCRAARTTNCKRTT (SEQ ID NO:2). The amplified fragment was sequenced and the amino acid sequence encoded was deduced. The complete nucleotide sequence and deduced amino acid sequence are shown in
The position of the aconitase C encoding gene in P. aeruginosa is compared to the location of the corresponding encoding sequence in S. typhimurium and E. coli in
Deduction of the aconitase C protein sequence, which contains 869 amino acids, permitted comparison using the BLAST similarity search program to known aconitases. The protein showed 61% similarity using this program to E. coli aconitase A, 60% similarity to P. aeruginosa aconitase A, 61% similarity to human IRP1, 53% similarity to human IRP2, and 41% similarity to pig mitochondrial aconitase.
The approximately 900 base pairs of the aconitase C insert obtained in Example 1 was amplified by PCR from the PA01 genome using the primers GTGGCACCGACAGCCATAC (SEQ ID NO:3) and GCGCCCGTCGAAGTTGCGGTT (SEQ ID NO:4). The amplified fragment was ligated into pBluescript-2 (KS+) (Stratagene) which had been cleaved with EcoRV and treated with tack DNA polymerase and dTTP to form intermediate plasmid pBSacnC. An approximately 1 kb DNA that encodes gentamicin resistance (GmR) was isolated from pUCGm described in Schweizer, H. D., Biotechniques (1993) 15:831–834 by digesting with SmaI. This amplified segment was cloned into the StuI site which resides in the acnC coding sequence in pBSanC to generate pBSacnCGm. The approximately 2 kb DNA fragment which contains the acnC sequence which was disrupted with the GmR cassette was isolated by treating this plasmid with HindIII and PstI and filled in with Klenow and dNTP's. This fragment was ligated to the SmaI ends of the conjugation plasmid pEX100T (Schweizer, H. D., et al., Gene (1995) 158:15–22) to obtain pEXacnCGm.
This plasmid was used to transform E. coli S17-1 and recombinant cells containing the plasmid were combined with an approximately equal amount of PA01 cells and plated on LB plates for conjugation. After conjugation, the bacteria were plated on Pseudomonas isolation agar plates containing Gm for selection of residue P. aeruginosa cells. Mutants for disruption of the acnC gene were further selected on PIA plates containing Gm and 4% sucrose. The successful transformants are designated PA01* or PA01-acaB:Gm.
The effect of acnC gene disruption on various virulent activities of PA01 were determined.
Exotoxin A production, determined by Western analysis of stationary phase supernatants did not show any change. However, hemolytic activity (tested by zone clearance on blood agar plates); proteolytic activity (tested by zone clearance on milk agar plates); elastase activity (tested by zone clearance on 2XYT-elastin plates); pyoverdin production (tested by pigment production on F agar plates); and pyocyanin production (tested by pigment production of P agar plates) were all diminished substantially.
When tested for growth on various carbon sources, no difference was observed when glucose, glutamate, citrate, isocitrate, succinate, acetate, pyruvate, butyrate, hexonate, or glyoxalate was used as a carbon source. However, although the wildtype could grow well on propionate, pentanate and the combination of glyoxalate and propionate as carbon sources, the acnC disrupted strain did not grow under these conditions.
The two strains were also tested for aconitase activity using the known conversion of aconitase citric acid to cis aconitic acid. PA01 and PA01* were grown in TSBD supplemented with 100 μM iron and the aconitase activity was determined at 6, 10 and 14 hours. PA01 exhibited 5, 15, and 19 units of activity per milligram of protein at these time points, respectively, while PA01* exhibited only 3, 10, and 16 units, respectively. PA01* also exhibited slightly reduced growth under these conditions.
Thus, disruption of the acnC gene completely abolishes growth in media where propionate is the carbon source, and diminishes growth in propionate-containing media with other carbon sources. Diminished aconitase activity in converting citrate to cis aconitate is also shown.
In addition, it has been shown that chemotaxis is affected and expression of several virulence factors is diminished.
In addition to PA01, an additional mucosal strain, FRD1 with gentamicin resistance was subjected to disruption of the acnC gene, as was a mucoid clinical isolate. For comparison, aconitase A was also similarly disrupted. All of the strains behaved similarly in respect to their ability to grow on propanediol in the presence and absence of propionic acid. All of the strains were able to grow in 1,2-propanediol, but were less able to grow using 1,3-propanediol as carbon source. Addition of propionic acid to the medium completely abolished the ability of both PA01 and FRD1 with disrupted aconitase C genes to grow under these conditions; disruption of the acnA gene did not result in this effect.
FRD1 is much more susceptible to propionate inhibition than wildtype PA01. While PA01 is able to grow on 0.4% propionate, inhibition of FRD1 growth occurs at levels of 0.1% propionate and inhibition is complete as low as 0.2% propionate in the medium.
It has also been noted that aconitase activity (measured as the conversion of citric acid to cis aconitate) is different in PA01 wildtype as compared to FRD1. While this activity was slightly higher in FRD1 after 4 hours of culture as compared to PA01, after 7 hours of culture the activity was statistically higher in PA01 cells and roughly equivalent after 14 hours.
It has also been shown that the results of propionate treatment can be obtained using odd-numbered chain fatty acids in the medium, as the metabolic products of these fatty acids include propionate. This is shown in Table 1 below which shows results of the addition of propionate, butyrate, hexanoic acid and pentanoic acid to M9 medium on wildtype PA01*, and PA01 with two irrelevant disruptions—acaA and icdI. As shown, both the presence of 0.4% propionate and 0.4% pentanoate disrupt the growth of the strain with a disrupted acnC gene, but these additions have no effect on the other strains tested.
The effect of the addition of 0.4% propionate to LB medium containing wildtype PA01, PA01*, and the more sensitive mucoid strain FRD1, was also tested. The wildtype does not respond to 0.4% propionate, but the modified strain PA01* containing a disrupted acnC gene does show diminished growth, as does the more propionate-sensitive strain FRD1.
FRD1, as set forth in Example 4, shows enhanced sensitivity to propionate, possibly due to reduced levels of TCA cycle activity, since propionic acid drains the TCA cycle intermediate oxalacetic acid.
This effect was tested by supplementing M9 minimal media with various TCA cycle components in the presence of 0.4% propionate. While glucose as a carbon source failed to reverse the negative effects on growth of propionate, the cell cycle intermediates acetate, aspartate, glutamate and malate were successful in doing so.
Ibuprofen is known to generate propionic acid when metabolized and was thus tested for its ability to inhibit the growth of FRD1. Other propionate-generating compounds include certain α4β1 integrin antagonists described by DuPlantier, A. J., et al., Bioorg. Med. Chem. Let. (2001) 11:2593–2596. Using absorbance at 600 nm as the criterion, growth inhibition was detected at a concentration of 0.5 mg/ml of ibuprofen added to 2 ml LB cultures; the percent inhibition at this level was 19%. Inhibition increased in a dose-dependent manner to 72% at 5 mg/ml, and was similar at 10 mg/ml (73%).
As mucoid production is necessary for attachment, the effect on mucoid production of addition of propionate can be tested by assessing the level of attachment of the cells to plastic tubes. This was measured as the absorbance at 570 nm of plastic tubes. Levels of as low as 0.1% propionate were able to diminish significantly the attachment of FRD1 cells to plastic tubes; the effect on attachment of PA01 was less dramatic.
The effect of various concentrations of propionate on the ability of antibiotics to curtail the growth of PA01 or FRD1 was tested. Levels of 0.2% propionate appeared to have no effect on PA01 growth in the presence of the antibiotics amikacin, carbenicillin, ciprofloxacin, tobramycin, and tetracycline. Growth was assured by the diameter of colonies formed in the presence of discs containing amikacin (AN-30), 30 μg, carbenicillin (CB-100), 100 μg, ciprofloxacin (CIP-5), 5 μg, tobramycin (NN-10), 10 μg, or tetracycline (TE-30), 30 μg. On the other hand, 0.2% propionate appeared to diminish the effect of CB-100 and CIP-5 on inhibiting growth of FRD1. There appears to be little, if any, effect on the efficiency of remaining antibiotics tested.